In the field of toxicology, high resolution assays now make it possible to discover differences in gene expression brought on by exposure to a particular xenobiotic. Such high-throughput, high-resolution molecular biology methods can be used to determine virtually all toxicant-induced changes in gene expression. A catalog of toxicant-induced gene expression changes would be useful to better predict animal toxicity in order to reduce costs, timelines and animal use by enhancing the probability that product candidates chosen for further development will pass regulatory testing requirements. Such a catalog would also enable scientists to better predict human toxicity, resulting in fewer compounds failing in clinical trials while better safeguarding human health.
The basis for these types of investigations is the expectation that toxicological endpoints (e.g. tumor formation) are the result of earlier molecular events. For example, by creating a catalog of changes in rat liver gene expression following treatment with phenobarbital, one can test whether early gene expression is as predictive as later readouts in assessing the nongenotoxic carcinogenicity of this compound in rats.
The power of transcriptional genomic analyses is that they can measure changes in the expression of thousands of genes, and a comprehensive catalog of expression changes can be envisioned. Using the same catalog of changes, other known nongenotoxic carcinogens (NGCs) could be assessed, as well as compounds known not to be NGCs in rats. Analysis of correlations between the changes and carcinogenesis, as well as analysis of the biological significance of the genes, should indicate whether there are specific genes or gene-expression patterns that predict carcinogenesis. Thus, there is a need in the art for catalogs or panels of predictive markers. Such panels of expressed genes would allow one to examine a greater number of candidate compounds in a shorter period of time prior to selecting a lead compound for traditional testing. As a result of this screening approach, the success rate of compounds in pre-clinical trials should improve dramatically.
These panels of predictive markers could also be used to assess the use of primary rat hepatocytes in high-throughput cell-based assays of toxicity and carcinogenicity. This would further increase the number of compounds that could be assessed, perhaps to the point where entire compound libraries could be assayed, and scores for potential toxicities could be created for each compound. Further, parallel analyses using both animal and human genes could be used to correlate the results from pre-clinical in vivo and in vitro data (using both cultured animal and cultured human cells) with human clinical data to create assays that better predict human toxicity.